The complexity of product assortments offered by online selling platforms makes the selection of appropriate items a challenging task. Customers can differ significantly in their expertise and level of knowledge regarding such product assortments. Consequently, intelligent recommender systems are required which provide personalized dialogues supporting the customer in the product selection process. In this paper we present the domainindependent, knowledge-based recommender environment CWAdvisor which assists users by guaranteeing the consistency and appropriateness of solutions, by identifying additional selling opportunities, and by providing explanations for solutions. Using examples from different application domains, we show how model-based diagnosis, personalization, and intuitive knowledge acquisition techniques support the effective implementation of customer-oriented sales dialogues. In this context, we report our experiences gained in industrial projects and present an evaluation of successfully deployed recommender applications.

Knowledge-based recommender technologies provide a couple of mechanisms for improving the accessibility of product assortments for customers, e.g., in situations where no solution can be found for a given set of customer requirements, the recommender application calculates a set of repair actions which can guarantee the identification of a solution. Further examples for such mechanisms are explanations or product comparisons. All these mechanisms have a certain effect on the behavior of customers interacting with a recommender application. In this paper we present results from a user study, which focused on the analysis of effects of different recommendation mechanisms on the overall customer acceptance of recommender technologies.

Abstract. In this paper we propose the introduction of dynamic logic programming – an extension of answer set programming – in recommender systems, as a means for users to specify and update their models, with the purpose of enhancing recommendations. 1

Abstract not found

We present ISeller an industry-strength recommendation system for online shopping platforms. The system supports several recommendation paradigms like collaborative, content-based and knowledge-based filtering as well as one-shot and conversational interaction modes. A generic user modelling component allows different forms of hybrid reasoning strategies as well as enables the system to support a process-oriented way of interactive selling in various product domains. This application paper contributes a comprehensive scenario for interactive selling on commercial platforms and thus motivates central user modelling services for recommendation systems. Furthermore, we give an outline on the technical architecture and the implemented system. Our presentation will be illustrated with actually fielded examples from domains of luxury-articles like coffee and cigars.

The effective integration of configuration sys-tem development with industrial software development is crucial for a successful implementation of a Mass Customization strategy. On the one hand, configuration knowledge bases must be easy to develop and maintain due to continuously changing product assortments. On the other hand, flexible integrations into existing enterprise applications, e-marketplaces and different facets of supply chain settings must be supported. This paper shows how the Model Driven Architecture (MDA) as an industrial framework for model development and interchange can serve as a foundation for standardized configuration know-ledge representation, thus enabling knowledge sharing in heterogeneous environments. Using UML/OCL as stan-dard configuration knowledge representation languages, the representation of configuration domain-specific modeling concepts within MDA is shown and a formal semantics for these concepts is provided which allows a common understanding and interpretation of configuration task descriptions.

This paper summarizes our experiences gained from several industrial advisory applications that were developed with the knowledge-based ADVISOR SUITE framework over the last years and gives an outlook on future extensions of the presented system. In the ‘experiences ’ section of the paper, we first address aspects related to the development of such applications, such as knowledge engineering, software maintenance, or testing. In addition, we describe the main requirements for such an advisory application to be perceived as an intelligent, value-adding service by the end users and finally summarize the results of an industrial study on how advisory applications are able to influence the buying behavior of online shoppers. The second part of the paper discusses current and future extensions of our system. The main lines of research addressed in this section are ‘Extended debugging support’, ‘Automated extraction of product data from web sources’, ‘Log mining and advanced data analysis’, and ‘Community-adapted advisory systems’. 1

Abstract not found

Abstract not found

Knowledge-based recommender systems are Web-based applications that exploit deep domain knowledge for generating buying proposals that match the individual needs and requirements of an online user. As in many domains the detailed customer requirements have to be elicited in an interactive dialog before the recommendation can be made, the development and in particular also the maintenance of the dynamic Web pages that form this personalized dialog are critical tasks, mostly because of the typically strong interdependencies between the recommendation and personalization knowledge. In this paper, we present ADVISOR SUITE, an integrated, domain-independent environment for the development of highly-interactive, personalized recommender applications. The main pillars of the presented system are a) an integrated, model-driven approach for designing all the required recommendation-, personalization- and interaction knowledge, and b) a mechanism that allows for the automatic generation of Web applications, which is of particular importance in prototyping-based, evolutionary development approaches. On the basis of the experiences we have made with the system in several industrial projects, we finally summarize key criteria and best practices of how to efficiently develop high-quality recommender applications with ADVISOR SUITE.